2020 Volume 43 Issue 4 Pages 693-696
In Japan, there is no national surveillance study of Clostridioides (Clostridium) difficile infection (CDI), and details about the epidemiology and treatment status of CDI are unknown. Additionally, clinical practice guidelines (CPGs) for CDI are published by four different institutions. All CPGs recommend that the antimicrobials, vancomycin (VCM) and metronidazole (MNZ), should be selected according to disease severity. However, the trends for VCM and MNZ use in Japan remain unclear. Therefore, this study was aimed at clarifying the secular trends for VCM and MNZ use based on sales data from 2006 to 2015 and discussing its impact on CDI status and drug costs. This is the first study to clarify the antibiotic use trends for CDI treatment. We found that the total use increased over time (r = 0.0013, Pfor trend < 0.0001). While VCM use significantly decreased (r = −0.0003, Pfor trend = 0.0002), MNZ use increased (r = 0.0017, Pfor trend < 0.0001). These results show that although treatment for CDI was in line with CPGs, CDI incidence might be on an increasing trend. Additionally, despite the increased total use, the total drug costs decreased by 55% ($ 25 million) from 2006 to 2015. It was also surmised that CDI treatment in compliance with CPGs would lead to a reduction in drug costs. Hence, to understand the epidemiology of CDI, it is important to continuously investigate the use of drugs used for CDI therapy.
Clostridioides (Clostridium) difficile (CD) infection (CDI) typically develops due to the disruption of normal intestinal flora because of frequent antibiotic use. The incidence of CDI is considered to be closely related to antibiotic use.1) Overseas, the increase in CDI incidence, severity, and the resultant mortality due to the emergence of highly toxic strains has become a significant problem.2,3) Therefore, in several countries, beginning with the United States, a national surveillance study of CDI epidemiology has been performed.4,5) However, thus far, no national surveillance study of CDI has been conducted in Japan. Therefore, the details regarding the epidemiology and treatment status of CDI in the country are unknown.
Oral vancomycin (VCM) and metronidazole (MNZ) have been recommended for CDI treatment.6) Additionally, it has been reported that VCM is more effective when compared to MNZ for treating severe and recurrent CDI.7) Clinical practice guidelines (CPGs) for CDI treatment established by the following four agencies are widely adopted: the Health Protection Agency and Department of Health (HPA/DH 2008), the European Society of Clinical Microbiology and Infectious Diseases (ESCMID 2009), the Society for Healthcare Epidemiology of America and Infectious Disease Society of America (SHEA/IDSA 2010), and the American College of Gastroenterology (ACG 2013). All CPGs recommend oral MNZ as the first-line drug for treating non-severe CDI and oral VCM for severe or recurrent CDI. Additionally, the CPGs recommend that VCM dosage should be varied depending on CDI severity. MNZ was an off-label drug for CDI in Japan until August 2012 and its price remains lower than that of VCM. Therefore, it is thought that clarifying the trends for VCM and MNZ use can be estimated based on CDI incidence, treatment status, and drug costs with time. However, the trends remain unclear in Japan. Hence, the aim of this study was to clarify the trends for VCM and MNZ use from 2006 to 2015 based on sales data and to discuss the impact on CDI status and drug costs.
Sales data for oral VCM and MNZ were obtained from IQVIA Services Japan K.K. (Tokyo, Japan). The surveillance period extended from 2006 to 2015. These data were collected using pharmaceutical sales distribution channels through wholesalers. The data were analyzed according to the Anatomical Therapeutic Chemical (ATC) classification by using defined daily dose (DDD) as a measurement unit.8) DDDs is defined as drug use divided by DDD. The 2019 version of the ATC/DDD system was applied to all of the data. The sales data were reported as DDD per 1000 inhabitants per day (DID), and the following equation was used:
 |
DDD was defined as 2 for VCM and 2 for MNZ. The population data were obtained from the Statistics Bureau of Japan.9)
Drug prices were obtained from services that provide drug price information reported in September 2019.10) The exchange rate was set at 100 Japanese yen per U.S. dollar. The prices of original and generic VCM were $25.366 and $11.131, respectively; the corresponding prices for MNZ were $0.355 and $0.193.
Statistical Analysis and Ethical ConsiderationThe trends for DID and drug costs for VCM and MNZ were evaluated by performing a linear regression. A two-sided value of p < 0.05 was considered statistically significant for all analyses. The need for informed consent was waived because anonymized sales data were used.
The trends for VCM and MNZ (in DID) from 2006 to 2015 are presented in Fig. 1a. Over time, the total drug use increased (r = 0.0013, Pfor trend < 0.0001, 95% confidence interval (CI): 0.0010 to 0.0016). VCM use significantly decreased (r = −0.0003, Pfor trend = 0.0002, 95% CI: −0.0004 to −0.0002) while MNZ use significantly increased (r = 0.0017, Pfor trend < 0.0001, 95% CI: 0.0013 to 0.0021).
(a) DID of MNZ and VCM, (b) Costs of MNZ and VCM. The values show DID (DDDs/1000 inhabitants/d) or costs in dollars ($ in millions). VCM, vancomycin; MNZ, metronidazole.
Trends for VCM and MNZ costs from 2006 to 2015 are presented in Fig. 1b. The total cost significantly reduced (r = −2.63, Pfor trend < 0.0001, 95% CI: −3.08 to −2.17) by approximately 55% ($ 25 million) from 2006 to 2015: VCM cost significantly reduced (r = −2.85, Pfor trend < 0.0001, 95% CI: −3.30 to −2.39) by approximately 60% ($ 27 million) and MNZ cost increased (r = 0.22, Pfor trend < 0.0001, 95% CI: 0.17 to 0.26) significantly by approximately 250% ($ 2 million).
To our knowledge, this is the first study to clarify the detailed trends for drugs used in CDI therapy in Japan from 2006 to 2015. While VCM use significantly decreased over time, MNZ use increased. Thus, total VCM and MNZ use increased over time. In addition, despite the increased total use, the total drug costs decreased by 55% ($ 25 million). It can be inferred that drug costs can be reduced by selecting antimicrobials recommended by CPGs.
The use of broad-spectrum antibiotics is one of the risk factors for CDI development and recurrence,11,12) and the incidence of CDI is included as one of the useful indicators of antimicrobial stewardship.13) Therefore, it is important to understand the incidence of CDI in Japan. However, a nationwide surveillance system has not yet been established in Japan. A systematic review of CDI epidemiology in Japan performed with relevant publications from January 2006 to November 2017 showed that the incidence rate of CDI was 0.8–4.7/10000 patient-days.14) However, because of some medical institutions and heterogeneity in patient populations, the trends for CDI incidence over time are unknown. In this study, the total VCM and MNZ use for CDI treatment significantly increased over time. Although it is an indirect statistic based on sales data, this finding indicates the possibility that CDI incidence has increased.
MNZ use significantly increased by 2.5-fold from 2006 to 2015. In Japan, MNZ was used as an off-label drug for CDI and anaerobic infection until August 2012. However, in other countries, the effect of MNZ for CDI has been reported since the 1980s,15,16) and the Ministry of Health, Labour and Welfare recommended MNZ as the first-line drug for moderate CDI in 2008.17) Therefore, it was assumed that there were many cases in which MNZ was used off-label until MNZ was covered by insurance. Additionally, MNZ use was further promoted in line with CPGs after the insurance application. However, the increased use of MNZ suggests increased high-dose and/or long-term use. It has been reported that side effects of MNZ, such as central nervous system disorders and gastrointestinal disorders are likely to develop with high doses and/or prolonged treatment.18) Thus, it is necessary to pay attention to these side effects. On the other hand, VCM use significantly decreased by 40% from 2006 to 2015. These findings may be due to the choice of MNZ or VCM depending on severity and a decrease in the daily dose of VCM from 2 g per day to 0.5 g per day.
Recently, in Japan, medical costs have exceeded 40 trillion Yen due to the declining birthrate and aging population, which has become a social problem.19) Increased medical costs have been attributed to drug costs.20) Since antimicrobial stewardship promotion is expected to reduce medical costs,21) we revealed the trend for VCM and MNZ costs. As a result, while MNZ costs increased by approximately 250% ($ 2 million), VCM costs decreased by approximately 60% ($ 27 million). Accordingly, the overall drug costs decreased by approximately 55% ($ 25 million). It was speculated that CDI treatment in compliance with CPGs would lead to a reduction in drug costs.
This study has some limitations. First, the data used are based on sales volume not actual administered volume. In addition, the background characteristics of the patient who used the antibiotics, the purpose of use, the number of patients administered, and the day of therapy are unknown. Second, there is the possibility that information such as that being sold directly to medical institutions is not included, but the sales data used accounts for approximately 98% of the total sales in Japan. Third, the status of CDI treatment is not accurately reflected because VCM and MNZ are not used exclusively for CDI treatment. Fourth, because DDD used to calculate DID is the assumed average maintenance dose per day for a drug used for its main indication in adults, it may be underestimated as the selection pressure. However, the trends of CDI therapy in Japan, which have been newly clarified in this research, will be useful in understanding the current status of epidemiology and treatment of CDI. In future, it will be easier to estimate the accurate CDI burden if data such as those presented in this study are compared with those obtained by surveillance based on the examination of microorganisms. Furthermore, detailed analyses using a national database based on health insurance claims data are necessary and it is important to establish a nationwide surveillance system for the incidence of CDI.
In Japan, bezlotoxumab, the first monoclonal antibody for CDI prophylaxis, and fidaxomicin, a new CDI treatment, were approved in November 2017 and July 2018, respectively. Additionally, in October 2018, the “Clostridioides (Clostridium) difficile infection treatment guideline” was first published by the Japanese Society of Chemotherapy and the Japanese Association for Infectious Diseases. In the future, CDI treatment in Japan is expected to change. It is important to continue to investigate the trends for various antimicrobials in Japan to clarify the impact of guidelines published in Japan. These results are expected to provide basic knowledge for evaluating antimicrobial stewardship.
The trends for antibiotic use for CDI treatment based on sales data from 2006 to 2015 were clarified for the first time in this study. Based on the findings, it is speculated that appropriate CDI treatment has been promoted in Japan in compliance with CPGs.
This work was supported by the Ministry of Health, Labour and Welfare (Grant numbers H28-shinkougyousei-ippan-004 and H31-shinkougyousei-shitei-005).
SI and YM designed the study. SI collected and analyzed the data. The manuscript was initially written by SI, and then critically reviewed by YM and RI. All authors have read the submitted manuscript and given their final approval for its publication.
The authors declare no conflict of interest.
